Hundreds of studies link poor marital functioning to depression. Few recent studies have investigated their association among adults in the second half of the lifespan. There is a need to understand how key marital variables affect wellbeing during midlife and beyond given the recent rise in divorce among the baby boomers. However, the causal, as opposed to merely correlational, pathway between marital functioning and depression is difficult to establish decisively. Twin methodology offers one way to control for nonrandom genetic and environmental selection confounds that often obscure causal findings in nonexperimental data. The proposed research will examine: whether changes in marital behaviors cause changes in depression over time in midlife (Specific Aim 1); whether midlife parenting transitions moderate the genetic risk for depressive symptoms associated with changes in marital dysfunction (Specific Aim 2); and whether the structure of facets of married life changes over different stages of midlife development (Specific Aim 3). The proposed research will use a genetically informed subsample of middle-aged married parents (N = 2,402) from the Midlife Development in the United States (MIDUS) study, a two-wave data set that includes multiple scales of marital functioning and clinical and nonclinical depressive symptom measures. The research training program is designed to develop the applicants methodological, quantitative, and theoretical skills to launch his program of research on depression and intimate relationships over the lifespan. The applicant will receive mentoring from theoretical and quantitative experts in his field to use latent stable and change score structural equation methods, gene- environment interaction methods, and other factor analytic methods (idiographic filtering) to study the etiological pathways underlying marital quality and depression during midlife, to operationalize and test stressful midlife moderators that may turn on the genetic ris factors mediating the association between depression and marital dysfunction, and to explore how some marital behaviors matter more during one stage of midlife than at others. While funded by Institutional NRSA Quantitative Training Grant (T32AG020500, PI Nesselroade), preliminary factor analyses with the MIDUS data reveal a genetic etiology for latent marital quality and depression, significant negative correlations between marital quality and depression within and over time, and sufficient power to detect small-to-moderate sized genetic and environmental main effects of marital quality on depression. The proposed research supports the NIAs Strategic Direction A-1 by helping map the social and genetic pathways between dysfunctional marriages and wellbeing. The findings will inform interventions for healthy physical and mental wellbeing in the second half of the lifespan.